A Fuzzy Logic Approach to Measure Complexity of Generic AspectOriented Systems

Aspect-oriented programming (AOP) is an emerging technique that provides a mechanism to clearly encapsulate and implement concerns that crosscut other modules. It is claimed that this technique improves code modularization and therefore reduces complexity of object-oriented programs (OOP). Most of the proposed complexity measurement frameworks for AOP are for AspectJ programming language. In this paper, a generalized framework for assessment of complexity of aspect-oriented (AO) systems, has been defined that takes into account three, the most well known families of available AOP languages, AspectJ, CaesarJ and Hyper/J. In order to automate complexity measurement, a tool has been developed using fuzzy logic, in which some set of rules have been defined as rule base. Using this tool, complexity of majority of AOP languages can be measured, which will further help in the measurement of external software qualities, such as maintainability, reusability, adaptability and understandability.